Date Issue Brief # ISSUE BRIEF
The Role of Technology Policies in Climate Mitigation Carolyn Fischer July 2009 Issue Brief #09‐08 Resources for the Future Resources for the Future is an independent, nonpartisan think tank that, through its social science research, enables policymakers and stakeholders to make better, more informed decisions about energy, environmental, natural resource, and public health issues. Headquartered in Washington, DC, its research scope comprises programs in nations around the world. PAGE 1 FISCHER | RESOURCES FOR THE FUTURE The Role of Technology Policies in Climate Mitigation Carolyn Fischer1 Stabilizing global greenhouse gas (GHG) concentrations at levels to avoid significant climate risks will require massive “decarbonization” of all the major economies over the next few decades. Additional activities will be required to reduce emissions from other GHGs and to promote carbon sequestration through soil management, afforestation, and avoided deforestation. Achieving the necessary scale of emissions reductions will require a multifaceted policy effort to support a broad array of technological and behavioral changes. And change on this scale will require sound, well‐thought out strategies; here I outline some core principles for guiding the design of clean technology policies, with a focus on energy. First off, pick winning technology policies. Many studies analyze the technological options for achieving deep reductions in GHG emissions. In a well‐known Science article, Pacala and Socolow (2004) introduced a now‐popular tool illustrating the “wedges” of potential reductions from available technologies to bring the emissions path to a stabilization target. These kinds of studies are informative, but they focus on the capacity of technologies, rather than the cost‐effectiveness of reduction options, the possibilities for innovation over time, or the role of policies in getting there. Economists who model climate policies, on the other hand, tend to focus on cost‐effective solutions, but often with less technological detail. All models have difficulty incorporating realistic representations of technological change, uncertainties, barriers, and nonmarket‐based policies. It is important to remember that energy projections are difficult for proven technologies, and even trickier for emerging ones. In one word, a key challenge for meeting emissions and technology goals is uncertainty. We are not sure what emissions reductions will ultimately be needed or what the corresponding prices will be. We do not necessarily have a good idea of the costs of large‐scale deployment of existing technologies, when breakthrough technologies might arrive, or to what degree the costs and quality of existing technologies will be improved. These kinds of uncertainties can create a tension—how to choose among them? On the one hand, policies should be as neutral as possible, to allow a broad range of technologies to emerge and …………………………………. 1
Carolyn Fischer is a senior fellow at Resources for the Future, 1616 P Street, Washington, DC 20036 USA. The author is grateful for helpful comments from Alex Long, Dave Sawyer, Thomas Sterner, David Popp, and Knut Einar Rosendahl. PAGE 1 FISCHER | RESOURCES FOR THE FUTURE compete, and to avoid the problem of governments attempting to pick winners. On the other hand, we cannot be fully neutral, given that we are largely aware of the major technological options that will be available over the coming decades and some technologies have specific barriers and potentials that may require targeted assistance. Carbon pricing is a technology policy The most important technology‐neutral policy and the core of any cost‐effective approach must be a strong and increasing price signal across the entire economy that carbon emissions are costly. Emissions pricing can be implemented either through a carbon tax or a broad‐based cap‐and‐trade system. The reason for a primary reliance on carbon pricing is twofold. First, technologies are only useful if people want to use them. While social values may influence some to become early adopters of hybrid cars or compact‐fluorescent light bulbs, financial self‐interest is the primary driver for most participants in a market economy. Carbon pricing makes clean technologies more cost‐competitive and provides “market pull” by encouraging their adoption. It also reduces some of the need for reliance on public innovation programs targeted specifically toward clean energy, as the market has more incentive to contribute. Furthermore, carbon pricing ensures that public spending on “market push” strategies of research, development, and deployment ultimately has greater impact, by increasing demand for these technologies.2 Second, many options are available for reducing emissions. Not only is there a huge array of technological solutions for electricity generation, production processes, building materials, and consumer appliances, but a variety of behavioral changes can contribute to smaller emissions footprints. No command‐and‐control regulation could efficiently prescribe all the appropriate activities that should be undertaken. Carbon pricing, on the other hand, creates incentives to do all these things: use less carbon‐intensive fuels and products, conserve energy, and develop and deploy emissions‐reducing technologies. Furthermore, when cost‐effective reductions are taken in the near term with current technologies, some pressure is lifted on the speed and depth of technological change needed in the future to reach a long‐term cumulative emissions goal.3 Technological change and turnover will be essential for deep reductions; however, a lack of emissions pricing is not the only roadblock. There are a host of other impediments to a robust market for clean technology RD&D: financial, regulatory, behavioral, and network barriers; knowledge and innovation spillovers; and scale economies, among other challenges. Moreover, political realities may constrain the carbon price from being high enough to induce the necessary transformation and innovation. While many experts agree that a carbon price is necessary, few believe that a carbon price alone is sufficient to achieve these goals cost effectively. A carbon price should be supported by complementary policies to address barriers to technological development and deployment. …………………………………. 2
For a broader discussion of the interaction between emissions pricing, spillovers, and public support for environmentally friendly technologies, see Fischer (2008). 3
Fischer and Newell (2008) show that, even with knowledge spillovers, policy cost‐effectiveness depends largely on the degree to which all options for reducing emissions are encouraged. While emissions pricing is the single most effective policy, an optimal portfolio also includes R&D support, achieving emissions reductions at significantly lower cost than any single policy. PAGE 2 FISCHER | RESOURCES FOR THE FUTURE In a sense, a carbon price addresses the primary barrier, which has been the lack of financial reward for climate‐friendly behavior and technologies. However, additional barriers or market failures may require additional policy tools, and many of these need not target specific technologies. Certain barriers lend themselves to broad and neutral policies The social value of research and innovation often surpasses what the innovators themselves can appropriate. These knowledge “spillovers” represent a kind of market failure, because by receiving only a fraction of the benefits, innovators have only a fraction of the incentive to engage in R&D. Studies of commercial innovations suggest that, on average, less than half of the gains to R&D return to the originator, although appropriation rates vary considerably over different types of innovations.4 Basic research, in particular, is an excellent candidate for government support, as the commercial applications are often distant and unknown. Other technologies may become commercially viable, but only when the carbon price is high enough. Although greater stringency in climate policies may be expected in the future, patent lifetimes are still limited. Consequently, the appropriation rates for climate‐friendly technologies are likely to be relatively low, at least initially, and rising over time, meaning some extra support during the transition can help clean technology development (Gerlagh et al. 2008). Even commercial innovations have spillovers—however, it is important to remember that spillovers are not the exclusive domain of clean energy technologies. With a carbon price in place, tax breaks and other public incentives for reflecting the additional social value of R&D are most efficient when they are broad‐based. In addition to a carbon price, other policies can ensure that the allocation of private R&D better follows social (including environmental) values. For instance, distorting subsidies for fossil‐based energy should be removed. In non‐OECD countries, subsidies are primarily used to keep consumer prices artificially low, resulting in overconsumption. If major developing countries would wipe out all energy subsidies, global carbon dioxide (CO2) emissions could fall by four to five percent (IEA 2002). In OECD countries, however, most of these subsidies are for fossil‐fuel production; for example, in the United States, half of energy subsidies go to fossil fuels, compared to five percent for renewables (IEA 2006).5 Of course, beneficiaries of subsidies will resist reform. Therefore, removing subsidies may require a gradual phasing out (French coal subsidies were reduced in a 20‐year program); transitioning to less distortionary forms of assistance (the United States replaced agricultural commodity price supports with a direct income support program); and educating the public about the benefits to rally support (IEA 2002). Another kind of indirect subsidy is the lack of policy to reflect the cost of other environmental damages, besides GHG emissions. Regulating conventional air and water pollutants with market‐based mechanisms will also help improve market signals and make clean energy sources relatively more competitive to their fossil‐fuel counterparts. …………………………………. 4
See, for example, Jones and Williams (1998). Many of these subsidies take the form of preferential tax treatment, relative to other sectors. The oil and gas industries in the United States and Canada have benefited from such provisions as accelerated depreciation, the expensing of exploration and development costs, and other investment tax breaks; direct expenditures on infrastructure and R&D; and the incomplete capture of resource rents through royalties—many of which disproportionately support the development of the relatively dirty oil sands (Taylor et al. 2005). 5
PAGE 3 FISCHER | RESOURCES FOR THE FUTURE Inefficient regulations, on the other hand, can impede technical progress. Unnecessary legal and regulatory barriers that favor incumbents should be removed to allow for better competition. Unfortunately, some of the energy sectors most relevant for GHG reductions also involve highly concentrated, natural monopolies. For example, regulators of power generation, transmission, and delivery must keep an eye on the ability of new entrants to join and compete. Licensing, regulations, and interconnection procedures must be clear, not overly burdensome, and coordinated across jurisdictions, while allowing for appropriate oversight to balance potential tradeoffs in economic and environmental costs. Often, streamlining regulations need not be technology‐specific and can benefit all participants, not just new green entrants. New technologies may also require explicit new policies to create regulatory certainty. For example, the long‐term impacts of large‐scale carbon capture and sequestration (CCS) remain uncertain and relevant regulations, guidelines, and industry protocols are needed to assign liability and develop good practices. Barriers may be general in origin, but require more specific attention Information. For markets to function, they require not only good property rights and competition, but also information. Some product characteristics are easily observable, but others—like nutritional content or energy consumption rates—are not available or credible without government intervention. Improving the availability and visibility of information, product‐specific labels, credible reporting standards, and educational campaigns can allow better consumer and firm decisionmaking at lower costs. Standards. Still, perfect information may not be enough. Consumer uncertainty about energy prices and the quality and reliability of new technologies being offered can contribute to seemingly myopic behavior. Poor choices can also arise when those making decisions about energy‐using appliances and building features are not the same people as those using or paying for the energy, such as in landlord‐tenant relationships. Coping with short payback horizons and principal‐agent problems can require product‐specific interventions, such as building codes and standards for energy efficiency and fuel economy. While these standards are generally informed by technological options, they need not be prescriptive of particular ways to meet the standards. Indeed, they should be designed so as to allow cost‐effective alternatives and ongoing incentives for improvement. Financing. Risk and payback horizons also influence investment decisions; if the private perceptions of these factors do not align with the public ones, then policies may be needed to assist financing and manage risks for publicly desirable projects. Technologies for which capital costs are very large (such as nuclear power, hydropower, and CCS) are more likely to need preferential financing or guarantees to reduce private investment risks. Even wind generation has high capital costs relative to operating costs; however, the capacity can be expanded more incrementally and policies to guarantee profitable production prices have typically been used to reduce investment risk, rather than finance guarantees, although investment tax credits are also common. Ultimately, greater certainty about the carbon pricing policy will also help to reduce risks and raise returns for low‐carbon technologies, and financing interventions should focus on narrowing the discrepancy between private and public payback horizons. PAGE 4 FISCHER | RESOURCES FOR THE FUTURE Some barriers are specific to certain technologies Scale economies. Economies of scale are an issue for many new technologies. Until enough units have penetrated the market, production costs are high and support services are scarce. Policies to address this barrier can legitimately help some new technologies gain acceptance and get off the ground, but they should be careful to avoid extended support for uneconomic technologies. An example is hybrid vehicle tax credits in the United States, which phase out after a certain number of models are sold. Portfolio standards also become easier to meet (and credit prices fall) as scale economies are met. Networks and infrastructure. Some technological options require new infrastructure and support networks in order to function. However, private actors are reluctant to take on activities that supply public goods, and most would prefer to wait for someone else to do it. The resulting network externalities are an important cause of “path dependence” or “technological lock‐in,” and public intervention may be required to change paths. Important examples lie in the distribution of fuels for transport: biofuels, hydrogen, compressed natural gas or CNG, or plug‐in electric would require new fuel (or battery) distribution and storage equipment, as well as new vehicle engines. Here it may be costly to allow multiple new options and thereby difficult to avoid picking a winner, so the decision must be made deliberately. For costly network infrastructure investments, there is value to waiting for more information, in order to bet confidently on the technology. Some infrastructure investments for carbon‐free generation technologies may also have network externalities. Real‐time energy metering can allow for time‐of‐use pricing to better manage electricity demand, for example. Direct‐current lines in buildings could allow solar cells to power many devices without inverters. Upgrades to “smart grid” transmission technologies can facilitate the incorporation of distributed generation and intermittent renewable energy sources. However, many infrastructure investments—like transmission lines for remote renewable energy sources—are better viewed as an additional cost to developing more capacity in those resources, although there may be other barriers related to siting or entry. Expansion of nuclear generation would require central infrastructure in the form of a waste storage facility, which involves its own tradeoffs. Tradeoffs. Many technologies that reduce GHGs may instead cause other environmental damages and risks, such as nuclear generation, which creates radioactive waste and security concerns. Hydropower affects aquatic ecosystems, fish spawning, and cultural resource access rights. Battery waste involves toxic chemicals; transmission lines can disturb other land uses; most generation siting raises “not in my backyard” (NIMBY) issues—and the list goes on. Public assessment of the tradeoffs is needed before allowing broad deployment. These assessments are also related to the regulatory regime for deploying technologies, and assuring that regime is appropriate but not unnecessarily long or cumbersome. Certain technologies may deserve preferential treatment In addition to addressing important market failures and barriers, policymakers may want to direct extra attention and support to certain kinds of technologies that have special potential. Some examples of especially desirable technologies are those that expand options and reduce costs of reaching deep PAGE 5 FISCHER | RESOURCES FOR THE FUTURE reductions, those that may have additional spillover benefits at home, and those that may have spillover benefits abroad, further reducing global emissions and improving the likelihood of more globally stringent GHG agreements. Backstop technologies. As heavily emphasized in the Stern Review (2006), there is an important role for technology policies that focus on bringing down the costs of reducing carbon emissions. When the future emissions target is uncertain, as well as the costs of reaching potential targets, both R&D and early abatement activities can facilitate the adoption of more ambitious targets and thus help reduce the expected costs of future abatement, adaptation, and damages. However, certain kinds of R&D may also help to reduce the degree of uncertainty in these costs and thereby carry extra value. In the climate policy case, the national or societal marginal abatement cost curve represents a sequence of technological options, each more costly than its predecessor. “Backstop” technologies are a particular kind of option. Conceptually, a true backstop technology is free to be replicated at a large scale without scarcity constraints, meaning that marginal costs (though relatively high) do not increase much as capacity is expanded. The presence of backstop technologies helps to flatten out the upper portion of the overall marginal abatement curve, meaning that if stricter‐than‐expected emissions targets are necessary, carbon prices will not need to rise astronomically. In other words, if it turns out that climate change is even more serious than we think, and we need to step up emission reductions dramatically in the future, an affordable backstop that can be expanded to basically any scale would be invaluable. Therefore, given the uncertainty we face, there is an added value to bringing down the costs of technologies that help flatten the marginal abatement cost curve. Of course, another way to keep options open is by reducing emissions more aggressively in the near term. But if backstop technologies can keep costs lower in the worst‐case scenarios, expected long‐term costs are also lower, and that in turn reduces pressure to engage in deeper reductions in the near term. In terms of true backstop technologies, the most‐discussed candidates are carbon capture and storage, nuclear, and solar (and, theoretically, fusion). The flow of solar energy to earth is particularly large in comparison to societal needs. Each has the possibility of being utilized at large scales, though location (and risk management) could be a constraining factor. RD&D programs that can lower costs, expand capacities, and accelerate how rapidly these capacities can be tapped have an added insurance value, beyond the gains that would be realized at the expected levels of utilization laid out in roadmaps. Comparative advantage. Countries may have national R&D deployment policies, but the development of new technologies is a global effort. Consequently, there may be opportunities for coordination (or free‐
riding, for that matter) and for specialization. Technology‐oriented agreements can be aimed at knowledge sharing and coordination, research, development or demonstration, and even deployment.6 Such commitments can increase the technological effectiveness of an agreement over emissions reductions, although they are generally weak policies in terms of environmental effectiveness on their own. (Even at the international level, technology policies are complements to mitigation policies.) International …………………………………. 6
For a discussion of technology oriented agreements, see de Coninck et al. (2008). PAGE 6 FISCHER | RESOURCES FOR THE FUTURE agreements over technology standards can also be attractive from a competitiveness point of view, ensuring that trading partners have similar cost burdens. On the other hand, technologies might become a source of competitiveness—under different circumstances, some countries will enjoy a comparative advantage in certain technologies. In this case, not all countries will want to engage in the same R&D deployment portfolio, but rather specialize to some extent. For example, countries with large availability of geological sequestration sites may prefer to invest more in CCS innovation. Global spillovers. Technology spillovers do not respect borders either and can inform priorities for dealing with global pollutants like GHGs. In particular, technological advances that support international agreements and efforts have additional value beyond what is appropriated at home—for example, some technologies may have better potential to be adopted among emerging economies that lack direct carbon regulation. Indeed, the availability of low‐cost abatement opportunities may help encourage these countries ultimately to take on hard emissions targets. As a result, developed countries will want to engage not only in technology transfer agreements, but also R&D deployment efforts that are likely to produce technologies to be transferred.7 Summary and Options We should recognize that not all barriers to adoption are market failures. Cost, reliability and quality issues, and risk are all legitimate aspects that the market should be allowed to weigh in choosing cost‐effective technologies. Furthermore, R&D market failures are not exclusive to energy technologies, and once most energy‐related market failures are addressed (as through carbon pricing), society must be wary of crowding out other legitimate innovation efforts. As a result, the main tools for encouraging climate‐friendly technologies should be those that encourage the market to make good choices more generally: pricing carbon emissions and other environmental damages, removing distorting subsidies and barriers to competition, and supporting R&D broadly. Some technologies face particular barriers, requiring society to make a decision of whether to support them, committing to major infrastructure investments or environmental risks. Other technologies may merit extra support, because they offer insurance against the possible need for deeper reductions, or because they have greater potential for being adopted in other parts of the world. Several policy options are available to support technological development. Broad‐based policies include R&D tax credits, funding universities and research institutions, and other public support for research through competitive grant processes. Scale economies can be supported through tax breaks, subsidies, performance standards (including tradable ones), or market‐share mandates. While the latter two policies also create an implicit subsidy to the targeted technology (like renewable energy sources), paid for by the non‐preferred sources, they have the advantage of not only requiring no public outlays, but also naturally phasing themselves out as the new technology becomes cost‐competitive. …………………………………. 7
See also Popp (2008) for insights into technology transfer policies. PAGE 7 FISCHER | RESOURCES FOR THE FUTURE More specific policies are required to address particular market failures and barriers, including information requirements, energy efficiency standards, building codes, and the like. In these cases, policies will generally be more effective, the more closely they target the specific market failure, as opposed to a specific technology. Standards perform better when they are flexible rather than prescriptive in terms of how the goal must be achieved. Finally, for those technologies identified as being particularly desirable, some narrower R&D policies are available. Traditionally, most policies subsidize inputs to research, either through specific tax credits, grants or contracts, or directed research in publicly funded laboratories. If government lacks the expertise or impartiality, allocation of these research funds can also be outsourced to independent third‐party managers given specific mandates.8 Technology prizes, on the other hand, offer financial inducement to an output, such as being the first to develop a specific advance or the contestant having made the most progress by a deadline. Newell and Wilson (2005) indicate that such methods have been successful in the past and they could play a supportive role in climate policy, although attention should be paid to the design features, including the technological target, the size and nature of the prize, and the method for selecting the winner. International engagement is another component of technology policy. Recognizing that climate mitigation and technological advances are a global effort, countries can leverage their own R&D resources with international partnerships and agreements to encourage knowledge sharing and broaden the markets for new technologies. Ultimately, the biggest driver of technological adoption and change will be the mitigation policy, which determines the demand for those technologies. An additional advantage of emissions‐pricing policies is their ability to generate revenue, which can help fund the complementary technology programs. However, that is not to say that all or even a particular share of those revenues needs to be explicitly earmarked for technology programs. Indeed, just as technologies should compete in the marketplace for adoption, technology policies should compete in the budget among all the worthy causes. Supporting climate‐friendly R&D deployment is certainly one, but so are transitional assistance, adaptation, tax relief, foreign aid, and a host of other demands unrelated to climate, including other innovations. Priority should be given to policies that enhance overall economic efficiency—broad R&D support, removing distortions, addressing regulatory barriers, reducing tax burdens, improving information, and supporting fundamental research. Then policymakers can turn to more targeted programs, fully considering the benefits and the tradeoffs. …………………………………. 8
An example is the Ontario Centres of Excellence, which operate somewhat like a publicly funded venture capital firm. PAGE 8 FISCHER | RESOURCES FOR THE FUTURE References Fischer, Carolyn and Richard Newell. 2008. Environmental and Technology Policies for Climate Mitigation. Journal of Environmental Economics and Management 55(2): 142–162. Fischer, Carolyn. 2008. Emissions Pricing, Spillovers, and Public Investment in Environmentally Friendly Technologies. Energy Economics 30(2): 487–502. de Coninck, Heleen, Carolyn Fischer, Richard G. Newell, and Takahiro Ueno. 2008. International Technology‐
Oriented Agreements to Address Climate Change. Energy Policy 36(1): 335–356. Gerlagh, Reyer, Snorre Kverndokk, and Knut Einar Rosendahl. 2008. Linking Environmental and Innovation Policy. FEEM Working Paper No. 53.2008. International Energy Agency (IEA). 2002. Reforming Energy Subsidies. Paris, France: International Energy Agency and UNEP Division of Technology, Industry and Economics. ———. 2006). Carrots and Sticks: Taxing and Subsidising Energy. Economic Analysis Division Note on Energy Subsidies and Taxes. 17 January. Paris, France: International Energy Agency. Jones, Charles I. and John C. Williams. 1998. Measuring The Social Return To R&D. The Quarterly Journal of Economics 113(4): 1119–1135. Stern, Nicholas. 2006. The Economics of Climate Change: The Stern Review. Cambridge: Cambridge University Press. Taylor, Amy, Matthew Bramley, and Mark Winfield. 2005. Government Spending on Canada's Oil and Gas Industry: Undermining Canada's Kyoto Commitment. Pembina Institute Report. Alberta, Canada: Climate Action Network. Newell, Richard G. and Nathan E. Wilson. 2005. Technology Prizes for Climate Change Mitigation. Discussion paper 05‐33. Washington, DC: Resources for the Future. Pacala, S. and R. Socolow. 2004. Stabilization Wedges: Solving the Climate Problem for the Next 50 Years with Current Technologies. Science 305(5686): 968–972. Popp, David. 2008. International Technology Transfer for Climate Policy. Center for Policy Research Policy Brief 39. Syracuse, NY: Syracuse University PAGE 9 FISCHER | RESOURCES FOR THE FUTURE